US12012014B2 - Method and apparatus for optimizing battery management system - Google Patents

Method and apparatus for optimizing battery management system Download PDF

Info

Publication number
US12012014B2
US12012014B2 US17/172,106 US202117172106A US12012014B2 US 12012014 B2 US12012014 B2 US 12012014B2 US 202117172106 A US202117172106 A US 202117172106A US 12012014 B2 US12012014 B2 US 12012014B2
Authority
US
United States
Prior art keywords
target vehicle
battery
data
bms
related algorithm
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active, expires
Application number
US17/172,106
Other languages
English (en)
Other versions
US20220250505A1 (en
Inventor
Jiucai Zhang
Jun Wang
Chao Liu
Jin Shang
Qiang Ren
Ao Mei
Dadiao NING
Sichao GUO
Sheng HUI
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou Automobile Group Co Ltd
Original Assignee
Guangzhou Automobile Group Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangzhou Automobile Group Co Ltd filed Critical Guangzhou Automobile Group Co Ltd
Priority to US17/172,106 priority Critical patent/US12012014B2/en
Priority to CN202180004460.XA priority patent/CN115210590A/zh
Priority to PCT/CN2021/083798 priority patent/WO2022170671A1/fr
Publication of US20220250505A1 publication Critical patent/US20220250505A1/en
Application granted granted Critical
Publication of US12012014B2 publication Critical patent/US12012014B2/en
Active legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L58/00—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles
    • B60L58/10—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries
    • B60L58/16—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries responding to battery ageing, e.g. to the number of charging cycles or the state of health [SoH]
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L3/00—Electric devices on electrically-propelled vehicles for safety purposes; Monitoring operating variables, e.g. speed, deceleration or energy consumption
    • B60L3/12—Recording operating variables ; Monitoring of operating variables
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L58/00—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles
    • B60L58/10—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/371—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC] with remote indication, e.g. on external chargers
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/392—Determining battery ageing or deterioration, e.g. state of health
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L2240/00—Control parameters of input or output; Target parameters
    • B60L2240/70—Interactions with external data bases, e.g. traffic centres
    • B60L2240/72—Charging station selection relying on external data
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L2260/00—Operating Modes
    • B60L2260/40—Control modes
    • B60L2260/46—Control modes by self learning

Definitions

  • the present disclosure relates to the field of battery management, in particular to a method and apparatus for optimizing a battery management system (BMS).
  • BMS battery management system
  • BMS battery management system
  • PEVs plugged in electric vehicles
  • PHEVs hybrid electric vehicles
  • a method for optimizing a battery management system comprises: obtaining training data, wherein the training data comprises data from a target vehicle and data from auxiliary vehicles, and the auxiliary vehicles are vehicles mounted with a same or similar battery system as the target vehicle; optimizing a BMS related algorithm for the target vehicle by performing transfer learning based on the training data, wherein the BMS related algorithm comprises one or a combination of: a battery model, parameters for the battery model, a battery management algorithm, and parameters for the battery management algorithm.
  • a BMS related algorithm for the target vehicle before optimizing a BMS related algorithm for the target vehicle by performing transfer learning based on the training data, further comprising: pretraining an initial BMS related algorithm based on experimental data of fresh battery cells for the target vehicle; or, establishing a basic BMS related algorithm based on experimental data of fresh battery cells for the target vehicle, and optimizing the basic BMS related algorithm based on calibration data based on test vehicles to obtain the initial BMS related algorithm, wherein the initial BMS related algorithm comprises one or a combination of: an initial battery model, parameters for the initial battery model, an initial battery management algorithm, and parameters for the initial battery management algorithm.
  • the method before obtaining training data, further comprises one or a combination of: selecting the auxiliary vehicles based on similarity between the auxiliary vehicles and the target vehicle; determining weights for the auxiliary vehicles based on the similarity between the auxiliary vehicles and the target vehicle.
  • the similarity between the auxiliary vehicles and the target vehicle comprises one or a combination of: the similarity in usage, the similarity in battery configuration, wherein the similarity in usage includes one or a combination of: the similarity in driving pattern and the similarity in charging pattern.
  • the similarity in battery system comprises one or a combination of: the similarity in battery chemistry, the similarity in battery composition, the similarity in battery age.
  • optimizing a BMS related algorithm for the target vehicle by performing transfer learning based on the training data comprises: determining a source domain based on the data from the auxiliary vehicles, and determining a target domain based on the data from the target vehicle; optimizing the BMS related algorithm for the target vehicle by performing transfer learning based on the source domain and the target domain.
  • determining a source domain based on the data from the auxiliary vehicles comprises: combining the battery data d i s of the auxiliary vehicles based on weights for the auxiliary vehicles to obtain weighted d s ; determining a general BMS related algorithm c s based on the weighted d s , and determining the source domain based on the general BMS related algorithm c s .
  • optimizing the BMS related algorithm for the target vehicle by performing transfer learning based on the source domain and the target domain comprises: tuning the general BMS related algorithm c s by using the battery data d t of the target vehicle to derive the BMS related algorithm c t of the target vehicle which satisfies a preset condition.
  • the preset condition comprises one of: the BMS related algorithm c t of the target vehicle has a minimized prediction error; the BMS related algorithm c t of the target vehicle has a cross-validation accuracy above a predetermined threshold.
  • the method further comprising: performing active learning to reselect the auxiliary vehicles and/or determine weights for the auxiliary vehicles.
  • the method further comprises: programming the BMS related algorithm for the target vehicle to the target vehicle.
  • an apparatus for optimizing a battery management system includes a hardware processor and a memory, wherein the hardware processor is configured to execute program instructions stored in the memory to: obtain training data, wherein the training data includes data from a target vehicle and data from auxiliary vehicles, and the auxiliary vehicles are vehicles mounted with a same or similar battery system as the target vehicle; and optimize a BMS related algorithm for the target vehicle by performing transfer learning based on the training data, wherein the BMS related algorithm includes one or a combination of: a battery model, parameters for the battery model, a battery management algorithm, and parameters for the battery management algorithm.
  • the apparatus is a cloud server or cloud platform or is located inside a cloud server or cloud platform.
  • the hardware processor is further configured to execute program instructions stored in the memory to: pretrain an initial BMS related algorithm based on experimental data of fresh battery cells for the target vehicle; or, establish a basic BMS related algorithm based on experimental data of fresh battery cells for the target vehicle, and optimizing the basic BMS related algorithm based on calibration data based on test vehicles to obtain the initial BMS related algorithm, wherein the initial BMS related algorithm includes one or a combination of: an initial battery model, parameters for the initial battery model, an initial battery management algorithm, and parameters for the initial battery management algorithm.
  • the hardware processor is further configured to execute program instructions stored in the memory to perform one or a combination of: selecting the auxiliary vehicles based on similarity between the auxiliary vehicles and the target vehicle; determining weights for the auxiliary vehicles based on the similarity between the auxiliary vehicles and the target vehicle.
  • the similarity between the auxiliary vehicles and the target vehicle includes one or a combination of: the similarity in usage, the similarity in battery configuration, wherein the similarity in usage includes one or a combination of: the similarity in driving pattern and the similarity in charging pattern.
  • the hardware processor is configured to execute program instructions stored in the memory to: determine a source domain based on the data from the auxiliary vehicles, and determining a target domain based on the data from the target vehicle; and optimize the BMS related algorithm for the target vehicle by performing transfer learning based on the source domain and the target domain.
  • the hardware processor is further configured to execute program instructions stored in the memory to: program the BMS related algorithm for the target vehicle to the target vehicle.
  • a non-transitory computer readable storage medium includes program codes which, when executed by a computing device, cause the computing device to: obtain training data, wherein the training data includes data from a target vehicle and data from auxiliary vehicles, and the auxiliary vehicles are vehicles mounted with a same or similar battery system as the target vehicle; and optimize a BMS related algorithm for the target vehicle by performing transfer learning based on the training data, wherein the BMS related algorithm includes one or a combination of: a battery model, parameters for the battery model, a battery management algorithm, and parameters for the battery management algorithm.
  • FIG. 1 is a flow chart of a method for optimizing a battery management system according to an embodiment of the present disclosure
  • FIG. 2 is a schematic diagram showing the structure of an apparatus for optimizing a battery management system according to an embodiment of the present disclosure
  • FIG. 3 is a schematic diagram showing the architecture for the adaptive transfer learning approach to optimal and personalized BMS according to an embodiment of the present disclosure
  • FIG. 4 is a schematic diagram for transfer learning according to an embodiment of the present disclosure.
  • FIG. 5 is a flowchart of the method for optimizing the BMS algorithm in the steps 3-4 according to an embodiment of the present disclosure.
  • BMS battery management system
  • the design for BMS algorithms generally requires laboratory test data for batteries and calibration data obtained via a large number of vehicle running processes.
  • the laboratory test data of the batteries is generally used as a basis for establishing the BMS algorithm, and after a basic algorithm model is established, model parameters are adjusted and optimized by test and calibration performed on real vehicles.
  • a method for solving the data shortage problem during BMS algorithm design by using transfer learning is mainly proposed in the embodiments of the present disclosure.
  • the method utilizes actual working data of a type of vehicle in which the batteries are applied and the working data of other types of vehicles in which the batteries are applied to perform transfer learning optimization on the BMS algorithm, which reduces the need for calibration data of a sample vehicle when the BMS algorithm is designed, reduces the cycle and cost of development of vehicles, and better adapts the BMS algorithm to the usage habits of actual users of the vehicle.
  • the fundamentally new adaptive transfer learning battery management design methodology requires only limited battery data at the beginning of life and uses cloud computing and connected vehicle technology to enable battery management systems to apply previously learned information to novel situations.
  • Such a battery management system is safer, more functional, and adapting quickly to unforeseen circumstances, ensuring an optimal operation of the battery system to guarantee for durability and reliability through a system's field lifetime experience, and personalizing the individual PEV experience according to each PEV charging and driving patterns as well as road and environmental conditions while ensuring battery life.
  • a method for optimizing a battery management system is provided.
  • the method may be executed by any device having a computing capability, for example, a cloud server, a cloud platform, an on-vehicle processing device, etc. Since the process in the method for optimizing the BMS requires relatively large computing capability and resources, it is suggested to execute the method by a cloud server or platform.
  • FIG. 1 shows a flow chart of a method for optimizing a battery management system according to an embodiment of the present disclosure. As shown in FIG. 1 , the method for optimizing the battery management system includes the following operations S 102 and S 104 .
  • training data is obtained, wherein the training data includes data from a target vehicle and data from auxiliary vehicles, and the auxiliary vehicles are vehicles mounted with a same or similar battery system as the target vehicle.
  • the training data may be obtained by receiving the data reported (e.g., real-timely uploaded) by the target vehicle or the auxiliary vehicles.
  • a BMS related algorithm for the target vehicle is optimized by performing transfer learning based on the training data, wherein the BMS related algorithm includes one or a combination of: a battery model, parameters for the battery model, a battery management algorithm, and parameters for the battery management algorithm.
  • the method may further include: pretraining an initial BMS related algorithm based on experimental data of fresh battery cells for the target vehicle; or, establishing a basic BMS related algorithm based on experimental data of fresh battery cells for the target vehicle, and optimizing the basic BMS related algorithm based on calibration data based on test vehicles to obtain the initial BMS related algorithm, wherein the initial BMS related algorithm includes one or a combination of: an initial battery model, parameters for the initial battery model, an initial battery management algorithm, and parameters for the initial battery management algorithm.
  • the method may further include one or a combination of: selecting the auxiliary vehicles based on similarity between the auxiliary vehicles and the target vehicle; determining weights for the auxiliary vehicles based on the similarity between the auxiliary vehicles and the target vehicle.
  • the similarity between the auxiliary vehicles and the target vehicle may include one or a combination of: the similarity in usage, the similarity in battery configuration, wherein the similarity in usage includes one or a combination of: the similarity in driving pattern and the similarity in charging pattern.
  • the similarity in battery system may include one or a combination of: the similarity in battery chemistry, the similarity in battery composition, the similarity in battery age.
  • the similarity in battery composition includes one or a combination of: the number of battery cells connected in series or in parallel, and a cell capacity
  • the operation S 104 may include: determining a source domain based on the data from the auxiliary vehicles, and determining a target domain based on the data from the target vehicle; optimizing the BMS related algorithm for the target vehicle by performing transfer learning based on the source domain and the target domain.
  • determining a source domain based on the data from the auxiliary vehicles includes: combining the battery data d i s of the auxiliary vehicles based on weights for the auxiliary vehicles to obtain weighted d s ; determining a general BMS related algorithm c s based on the weighted d s , and determining the source domain based on the general BMS related algorithm c s .
  • optimizing the BMS related algorithm for the target vehicle by performing transfer learning based on the source domain and the target domain includes: tuning the general BMS related algorithm c s by using the battery data d t of the target vehicle to derive the BMS related algorithm c t of the target vehicle which satisfies a preset condition.
  • the preset condition includes one of: the BMS related algorithm c t of the target vehicle has a minimized prediction error; the BMS related algorithm c t of the target vehicle has a cross-validation accuracy above a predetermined threshold.
  • the method further including: performing active learning to reselect the auxiliary vehicles and/or determine weights for the auxiliary vehicles.
  • the method may further include: programming the BMS related algorithm for the target vehicle to the target vehicle.
  • FIG. 2 is a schematic diagram showing the structure of an apparatus for optimizing a battery management system according to an embodiment of the present disclosure.
  • the apparatus includes a hardware processor 22 and a memory 24 , wherein the hardware processor 22 is configured to execute program instructions stored in the memory 24 to perform the method described in the previous embodiment and exemplary embodiments.
  • the apparatus may be a cloud server or cloud platform or may be located inside a cloud server or cloud platform.
  • the hardware processor is further configured to execute program instructions stored in the memory to: pretrain an initial BMS related algorithm based on experimental data of fresh battery cells for the target vehicle; or, establish a basic BMS related algorithm based on experimental data of fresh battery cells for the target vehicle, and optimizing the basic BMS related algorithm based on calibration data based on test vehicles to obtain the initial BMS related algorithm, wherein the initial BMS related algorithm includes one or a combination of: an initial battery model, parameters for the initial battery model, an initial battery management algorithm, and parameters for the initial battery management algorithm.
  • the hardware processor is further configured to execute program instructions stored in the memory to perform one or a combination of: selecting the auxiliary vehicles based on similarity between the auxiliary vehicles and the target vehicle; determining weights for the auxiliary vehicles based on the similarity between the auxiliary vehicles and the target vehicle.
  • the similarity between the auxiliary vehicles and the target vehicle includes one or a combination of: the similarity in usage, the similarity in battery configuration, wherein the similarity in usage includes one or a combination of: the similarity in driving pattern and the similarity in charging pattern.
  • the hardware processor is configured to execute program instructions stored in the memory to: determine a source domain based on the data from the auxiliary vehicles, and determining a target domain based on the data from the target vehicle; and optimize the BMS related algorithm for the target vehicle by performing transfer learning based on the source domain and the target domain.
  • the hardware processor is further configured to execute program instructions stored in the memory to: program the BMS related algorithm for the target vehicle to the target vehicle.
  • a non-transitory computer readable storage medium includes program codes which, when executed by a computing device, cause the computing device to: obtain training data, wherein the training data includes data from a target vehicle and data from auxiliary vehicles, and the auxiliary vehicles are vehicles mounted with a same or similar battery system as the target vehicle; and optimize a BMS related algorithm for the target vehicle by performing transfer learning based on the training data, wherein the BMS related algorithm includes one or a combination of: a battery model, parameters for the battery model, a battery management algorithm, and parameters for the battery management algorithm.
  • BMS battery management system
  • Battery management methodology can be considered as a transfer learning problem.
  • FIG. 3 is a schematic diagram showing the architecture for the adaptive transfer learning approach to optimal and personalized BMS according to the embodiment of the present disclosure.
  • active learning optimally selects the most informative auxiliary vehicles
  • transfer learning makes use of training data from other auxiliary vehicles to derive battery model and algorithms.
  • the adaptive transfer learning approach is specifically described in detail as below.
  • an initial battery model, algorithms, and their parameters are developed based on the very limited battery data in laboratory and integrated into the battery management system for each new vehicle.
  • the transfer learning algorithm in the cloud will create an adaptive battery model, adaptive algorithms, and their corresponding parameters according to the field data.
  • the created adaptive battery model, algorithms, and their corresponding parameters will be programmed to each vehicle through the over-the-air programming.
  • the auxiliary vehicles contain the similar driving and charging patterns and thus we use transfer learning to transfer knowledge from the informative data of auxiliary vehicles to the target vehicle. Since the auxiliary vehicles are only a small portion of all PEVs, the active transfer learning will significantly reduce the data size to improve the transfer efficiency while achieving satisfactory performance.
  • the transfer learning algorithm will use transferred knowledge from auxiliary vehicles and the target vehicles to optimize battery model, algorithms, and their parameters. To personalize the vehicles performance, the knowledge from the auxiliary vehicles is weighted according to the target vehicle.
  • Transfer learning aims at boosting the learning process of updating battery model and battery management algorithms for different behaviors by transferring knowledge from an old battery to a new battery, from one battery chemistry to another one, or different driving profiles.
  • auxiliary and target vehicles, source and target domains, source and target tasks, and source and target conditions can vary in four ways, which we will illustrate in the following.
  • the feature spaces of the source and target domain are different.
  • the vehicle can be different types such as EV. HEV, and PHEV.
  • Each type of vehicle may own different types of batteries with unique battery chemistry and size.
  • Each battery may have unique battery life.
  • the battery data are different because each driver has a unique driver profile, and the vehicle may also have difference features.
  • the battery models and battery management algorithms to manage the battery behaviors are different due to different battery ages and chemistries.
  • FIG. 4 is a schematic diagram for transfer learning according to the embodiment of the present disclosure.
  • the size of line indicates the weights of transferred knowledge.
  • For each target vehicle we combine data from auxiliary vehicles for building battery model and battery management algorithms, where the contribution of the data from an auxiliary subject is determined by the response similarity between an auxiliary vehicle and the target vehicles.
  • the detailed implementation of the transfer learning algorithm is provided as follows.
  • Input for the transfer learning algorithm includes:
  • Output for the transfer learning algorithm includes:
  • the transfer learning uses data from auxiliary data to derive a general battery model architecture and initial parameters as well as battery management algorithms. After the target vehicles got this battery model architecture and initial parameters as well as battery management algorithms, it will refine the parameters and setpoints of the battery management algorithms to personalize them. Specifically, the battery model and battery management algorithms are firstly developed based on the experimental data, and then the battery model and battery management algorithms are trained by data from auxiliary vehicles and the target vehicle. The final battery model and battery management algorithm are a weighted voting from all the battery model and battery management algorithms. Herein, weighted voting can be the same or assigned based on the importance of the information.
  • BMS battery management system
  • the BMS algorithm design method generally requires laboratory test data for cells and calibration data based on a large number of vehicle travel processes.
  • the laboratory test data of the battery is generally used as a basis for establishing an algorithm, and after establishing a basic algorithm model, the model parameters are optimized through test calibration on an actual vehicle.
  • a common method is to mount a test vehicle of a corresponding vehicle model, and to test and calibrate relevant working conditions before sale of the vehicle.
  • a main problem is data shortage, for example, there is not enough field data such as working conditions, driving behavior and environment, and there is also not enough field data for the vehicle since as a new vehicle there is not so much actual field data. Through transfer learning, the data shortage problem for a new vehicle can be addressed.
  • the method for optimizing the battery management system generally includes the following four steps:
  • the data involved in the BMS algorithm design mainly includes data of four aspects:
  • FIG. 5 is a flowchart of the method for optimizing the BMS algorithm in the steps 3-4 according to an embodiment of the present disclosure.
  • auxiliary vehicles are selected from field data of various vehicles mounted with the battery and the proportional weights are input for each selected vehicle, and this step can be implemented according to the degree of similarity of the battery systems of different vehicles in the following aspects: the chemical system of the battery cells, the number of battery cells connected in series or parallel, the age of the battery cells, etc.
  • the BMS model and parameters including parameters such as battery charge state, open circuit voltage, internal resistance, terminal voltage, capacitance, etc.
  • the transfer learning method based on the field data of the selected vehicles.
  • the model accuracy satisfies the requirement
  • the iteration is stopped; otherwise, the vehicle selection method is optimized by using the active learning method, and the transfer learning optimization of the model is resumed after a new set of auxiliary vehicles is selected.
  • transfer learning of different life cycles of the same type of vehicle is conducted, i.e., the BMS algorithm is optimized based on real-time data in an entire life cycle of the vehicle, thereby solving the problem of shortage of test calibration data.
  • transfer learning between different vehicles is achieved, i.e., transfer optimization is performed on the BMS algorithm by using the field data of auxiliary vehicles mounted with the same battery system, thereby solving the problem that actual field data of the battery system cannot be obtained since there is no real vehicle to mount the battery system.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Sustainable Development (AREA)
  • Sustainable Energy (AREA)
  • Power Engineering (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Secondary Cells (AREA)
  • Electric Propulsion And Braking For Vehicles (AREA)
US17/172,106 2021-02-10 2021-02-10 Method and apparatus for optimizing battery management system Active 2043-04-05 US12012014B2 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
US17/172,106 US12012014B2 (en) 2021-02-10 2021-02-10 Method and apparatus for optimizing battery management system
CN202180004460.XA CN115210590A (zh) 2021-02-10 2021-03-30 优化电池管理系统的方法和设备
PCT/CN2021/083798 WO2022170671A1 (fr) 2021-02-10 2021-03-30 Procédé et dispositif d'optimisation d'un système de gestion de batterie

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US17/172,106 US12012014B2 (en) 2021-02-10 2021-02-10 Method and apparatus for optimizing battery management system

Publications (2)

Publication Number Publication Date
US20220250505A1 US20220250505A1 (en) 2022-08-11
US12012014B2 true US12012014B2 (en) 2024-06-18

Family

ID=82703543

Family Applications (1)

Application Number Title Priority Date Filing Date
US17/172,106 Active 2043-04-05 US12012014B2 (en) 2021-02-10 2021-02-10 Method and apparatus for optimizing battery management system

Country Status (3)

Country Link
US (1) US12012014B2 (fr)
CN (1) CN115210590A (fr)
WO (1) WO2022170671A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12012014B2 (en) * 2021-02-10 2024-06-18 Guangzhou Automobile Group Co., Ltd. Method and apparatus for optimizing battery management system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200101850A1 (en) * 2018-10-01 2020-04-02 Honda Motor Co., Ltd. Electric charge management system and method for a vehicle
US20210005027A1 (en) * 2019-07-04 2021-01-07 QodeLogix Systems Inc. System and method for battery maintenance management
US20210380013A1 (en) * 2020-06-07 2021-12-09 Blitz Electric Motors Ltd. Optimization of multiple battery management for electric vehicle fleets
US20220250505A1 (en) * 2021-02-10 2022-08-11 Guangzhou Automobile Group Co., Ltd. Method and Apparatus for Optimizing Battery Management System
US20220305945A1 (en) * 2019-08-27 2022-09-29 Janus Electric Pty Ltd Electric vehicle battery network management system, method and vehicle
US11609273B2 (en) * 2020-12-14 2023-03-21 Guangzhou Automobile Group Co., Ltd. Method and system for optimizing BMS model, storage medium and electric vehicle

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102959418B (zh) * 2010-06-24 2016-04-27 松下知识产权经营株式会社 获取电池的劣化度的方法和系统
KR102215450B1 (ko) * 2014-06-24 2021-02-15 삼성전자주식회사 배터리의 상태 정보를 학습 및 추정하는 장치 및 방법
WO2018113962A1 (fr) * 2016-12-21 2018-06-28 Volvo Truck Corporation Système de gestion de batterie et procédé de commande d'un système de gestion de batterie
CN107958136A (zh) * 2017-11-24 2018-04-24 广州市香港科大霍英东研究院 一种基于模型迁移的电池模型构建方法、系统及装置
KR102608468B1 (ko) * 2017-11-28 2023-12-01 삼성전자주식회사 배터리 상태 추정 방법 및 장치
CN108414937A (zh) * 2017-12-08 2018-08-17 国网北京市电力公司 充电电池荷电状态确定方法及装置
US11555858B2 (en) * 2019-02-25 2023-01-17 Toyota Research Institute, Inc. Systems, methods, and storage media for predicting a discharge profile of a battery pack
KR20190100114A (ko) * 2019-08-09 2019-08-28 엘지전자 주식회사 배터리 장치 및 그 제어 방법

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200101850A1 (en) * 2018-10-01 2020-04-02 Honda Motor Co., Ltd. Electric charge management system and method for a vehicle
US20210005027A1 (en) * 2019-07-04 2021-01-07 QodeLogix Systems Inc. System and method for battery maintenance management
US20220305945A1 (en) * 2019-08-27 2022-09-29 Janus Electric Pty Ltd Electric vehicle battery network management system, method and vehicle
US20210380013A1 (en) * 2020-06-07 2021-12-09 Blitz Electric Motors Ltd. Optimization of multiple battery management for electric vehicle fleets
US11609273B2 (en) * 2020-12-14 2023-03-21 Guangzhou Automobile Group Co., Ltd. Method and system for optimizing BMS model, storage medium and electric vehicle
US20220250505A1 (en) * 2021-02-10 2022-08-11 Guangzhou Automobile Group Co., Ltd. Method and Apparatus for Optimizing Battery Management System

Also Published As

Publication number Publication date
WO2022170671A1 (fr) 2022-08-18
CN115210590A (zh) 2022-10-18
US20220250505A1 (en) 2022-08-11

Similar Documents

Publication Publication Date Title
Li et al. Big data driven lithium-ion battery modeling method based on SDAE-ELM algorithm and data pre-processing technology
CN113095558B (zh) 一种智能网联汽车的迭代优化多尺度融合车速预测方法
EP4208725B1 (fr) Estimateur d'état de santé de batterie
US12000895B2 (en) Method for predicting an aging state of a battery
US11609273B2 (en) Method and system for optimizing BMS model, storage medium and electric vehicle
US20220373609A1 (en) State Value for Rechargeable Batteries
Shi et al. Electric vehicle battery remaining charging time estimation considering charging accuracy and charging profile prediction
Ermon et al. Learning policies for battery usage optimization in electric vehicles
CN114325440A (zh) 借助于机器学习方法运行用于为设备提供电能量存储器的预测老化状态的系统的方法和装置
CN112977412A (zh) 一种车辆控制方法、装置、设备及计算机存储介质
US12012014B2 (en) Method and apparatus for optimizing battery management system
Refaai et al. Battery energy forecasting in electric vehicle using deep residual neural network
CN113687237B (zh) 一种保障电气安全的锂电池剩余充电时间预测方法
CN112896171B (zh) 车辆的控制方法、装置、设备、车辆和存储介质
Valenti et al. Battery aging-aware online optimal control: An energy management system for hybrid electric vehicles supported by a bio-inspired velocity prediction
Meis et al. Guide for the Focused Utilization of Aging Models for Lithium-Ion Batteries-An Automotive Perspective
Jayaraman et al. Accurate state of charge prediction for lithium-ion batteries in electric vehicles using deep learning and dimensionality reduction
Mousaei et al. Machine learning-based regression models for state of charge estimation in hybrid electric vehicles: A review
US20240060786A1 (en) System for modelling energy consumption efficiency of an electric vehicle and a method thereof
US20240061971A1 (en) System for modelling energy consumption efficiency of an electric vehicle and a method thereof
CN113619447B (zh) 电动汽车电池荷电状态的预测方法
Gesner et al. Data-Enhanced Battery Simulator for Testing Electric Powertrains
US20230176137A1 (en) Method and system for determining a remaining useful lifetime of a battery
CN117863886A (zh) 增程车辆的续航预测方法、装置、电子设备及介质
Chun et al. Maximizing the Performance of a Lithium-Ion Battery Aging Estimator Using Reinforcement Learning

Legal Events

Date Code Title Description
FEPP Fee payment procedure

Free format text: ENTITY STATUS SET TO UNDISCOUNTED (ORIGINAL EVENT CODE: BIG.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

STPP Information on status: patent application and granting procedure in general

Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION

STPP Information on status: patent application and granting procedure in general

Free format text: NON FINAL ACTION MAILED

STPP Information on status: patent application and granting procedure in general

Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER

STPP Information on status: patent application and granting procedure in general

Free format text: NOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONS

STPP Information on status: patent application and granting procedure in general

Free format text: PUBLICATIONS -- ISSUE FEE PAYMENT RECEIVED

STPP Information on status: patent application and granting procedure in general

Free format text: PUBLICATIONS -- ISSUE FEE PAYMENT VERIFIED

STCF Information on status: patent grant

Free format text: PATENTED CASE